{"schemaVersion":1,"target":{"type":"reading","id":"c288de419a80fbea93bc6d54a25a4edf0a8c4c11da59c54b72a1c1e6739caf7d","revision":"ada9a6f1-5205-4ee0-879f-d0cf37ce1fce","revisionKind":"publication-id","title":"arXiv cs.AI: Leaderboard Hacking: Preference-Based Model Evaluations are Vulnerable to Manipulation","url":"https://escapemeridian.com/readings/arxiv-cs-ai-leaderboard-hacking-preference-based-model-c288de41"},"source":{"type":"source","id":"https://arxiv.org/abs/2611.14769","revision":"487e0a425ea8d2a824bfb519781a2d3f9fdba50b67c4ada3d0530cf865429069","revisionKind":"sha256-source-text","title":"Leaderboard Hacking: Preference-Based Model Evaluations are Vulnerable to Manipulation","url":"https://arxiv.org/abs/2611.14769"},"status":"available","items":[],"next":null,"scope":"Explicit citations in public readings and public file attachments to this reading. Shared topics are not citations. Other artifact types need an explicit reference adapter."}